Therapeutic options for patients diagnosed with glioblastoma multiforme (GBM) remain discouraging, despite the progress made in understanding the genetic lesions that cause this disease. In the last few years, drugs have been developed that target genetic alterations frequently found in GBMs. For example, the epidermal growth factor receptor (EGFR) is amplified and overexpressed in one third to one half of all GBMs. However, inhibitors of EGFR do not inhibit the growth of all tumors with EGFR overexpression, suggesting that additional genetic or biochemical alterations are important in determining response to these agents. We previously showed that activity of the phosphoinositide 3-kinase pathway is critical in determining the ability of GBMs to respond to these drugs - in tumors in which this pathway is high, no patient responded to EGFR inhibitors, a finding that has been independently corroborated. We are therefore initiating a phase 2 trial of an EGFR inhibitor (erlotinib) in recurrent GBM patients restricted to those that display low PI3-kinase signaling in the resected tumor. We will also further refine our ability to prospectively detect tumors that are most likely to respond to EGFR inhibitors by using human GBM xenografts in mice. Using a mouse model, we will also be able to test what components of PI3-kinase signaling appear to be most important in determining response to EGFR inhibitors. In the initial funding period, we demonstrated that low-grade gliomas show a high incidence of methylation at the PTEN promoter. PTEN is a tumor supressor that normally antagonizes PI3-kinase activity, and is frequently mutated in primary, but not secondary GBM tumors. We also showed that PTEN methylation was very frequent in secondary GBM tumors, suggesting that the PI3-kinase pathway is important in both primary and secondary GBMs, and that methylation of PTEN defines a distinct pathway that defines the low-grade to high-grade progression. We therefore propose to use an inhibitor of the PI3-kinase pathway, rapamycin, in recurrent, progressive, low-grade gliomas. Finally, we will also evaluate the activity of new PI3-kinase inhibitors, recently developed by the pharmaceutical industry, and assess how specific molecular features should guide the choice of targeted therapy for a particular glioma. These novel agents display promising activity as single agents, and in combination with both molecularly targeted therapeutics and conventional cytotoxic therapy.